Aliso: A Journal of Systematic and Evolutionary Botany

Volume 4 | Issue 2 Article 4

1959 Tenuis S Wats., A New Transfer to the Subsection Evansia Lee W. Lenz

Follow this and additional works at: http://scholarship.claremont.edu/aliso Part of the Botany Commons

Recommended Citation Lenz, Lee W. (1959) "Iris Tenuis S Wats., A New Transfer to the Subsection Evansia," Aliso: A Journal of Systematic and Evolutionary Botany: Vol. 4: Iss. 2, Article 4. Available at: http://scholarship.claremont.edu/aliso/vol4/iss2/4 ALISO

VoL. 4, No.2, pp. 311-319 ]UNE 26, 1959

IRIS TENUIS S. WATS., A NEW TRANSFER TO THE SUBSECTION EV ANSIA

LEE w. LENZ In northern Oregon, not far from Mount Hood in the Cascade Range, there is found a small iris which has generally been included within the series Californicae. This , Iris tenuis, occurs along the Clackamas River and its tributary, Eagle Creek, and so far as is known is endemic to Clackamas County where it is found growing in cool shady spots in moist soil or leaf mold. R. C. Foster (1937) and Dykes ( 1913) both cite sheets of a collected by L. F. Henderson in Washington County, Oregon, in 1884. The precise locality where these were gathered is not given and no recent collections from that county are known. The ecological habitat where I. tenuis is found is quite different from that occupied by any of the Californicae. As Starker (1935) wrote, the species is very often found among the dense undergrowth that covers the canyon floor. However, in another area, not further identified, he reports finding it in abundance on a gently sloping hillside quite free from underbrush except for a few rhododendrons and hazel bushes. Here the Douglas Fir trees were from three to four feet in diameter and "the iris covered the forest floor like grass on any lowland meadow." In this locality the plants were growing in leaf mold and decaying moss. That Iris tenuis does not rightfully belong with the Californicae has been suspected by numerous workers, and Dykes ( 1913) wrote that "it is at once distinguished from all other irises from Western America by its deeply forked stems." R. C. Foster (1937) likewise recognized it as foreign to the group but he did not remove it from the Californicae. Clarkson ( 195 5) stated, "it clearly does not belong in the sub­ section with the other members of the Californicae" and in his treatment of the Oregon species he excluded it from that group. The present author (Lenz, 1958) has likewise excluded it from the Californicae. Cytologically it has been shown (Simonet, 1934; R. C. Foster, 1937; Lenz, 1950; Clarkson, 1955; Smith and Clarkson, 1956) that the Californicae (sensu Clarkson, 195 5 ; Lenz, 1958) are all uniformly 2n = 40 whereas I. tenuis has been shown to be 2n=28 (Simonet, 1934; Smith and Clarkson, 1956). On the basis of the inclusion of I. tenuis within the Californicae the series would then appear to possess two basic chromosome numbers, 2n=40 and 2n=28. As pointed out by several investigators (Dykes, 1913; R. C. Foster, 1937; Lenz, 1958), the nearest relatives of the Cali­ fornicae are the Eurasian Sibiricae which also have been shown (Simonet, 1932, 1934) to have two basic chromosome numbers: I. delavayi, I. wilsonii, I. forrestii, I. chrysographes, I. bulleyana and I. clarkei with 2n=40, and I. sibirica and I. ori­ entalis with 2n=28. Mainly on the basis of parallelism in chromosome numbers Simonet (1934) united I. tenuis, I. sibirica, I. orientalis, and I. prismatica into the subsection Sibiricae with a basic chromosome number of x=7. Simonet then united the remaining members of the Californicae, i.e., all those species now recognized by Clarkson (1955) and Lenz (1958) as belonging to that series, with the 40 chromo­ some members of the Sibiricae into a new subsection, Chrysographes. As R. C. Foster

[311] 312 ALISO [VOL. 4, No.2

(1937) clearly pointed out, this alignment of the species almost completedly disre­ gards morphological characters and geographical distributions. Concerning the possible affiliations of I. tenuis with other members of the genus, R. C. Foster (1937) noted that the broad pale green leaves were much like those of a giant I. cristata, a member of the subsection Evansia. Clarkson (1955) likewise noted certain morphological similarities between the two but he was impressed with the differences in chromosome numbers between the two, I. cristata having 2n=36 compared with I. tenuis with 2n=28. More recently Clarkson (1958) has erected a new subsection, the Oregonae, for this species, saying only, "Regardless of origin, I. tenuis is sufficiently distinct, morphologically, cytologically, and geographically, to warrant erection of a new subsection which is accordingly proposed." Clarkson (1959) subsequently reduced the category from section to series because Lawrence ( 195 3) had earlier pointed out that Apogon as a sectional name is invalid and cannot be used but that it was valid as a subsection name as used by Bentham. Thus Lawrence changed the subdivisions of section Apogon to series of subsection Apogon. Apparently none of these workers noted the similarities between I. tenuis and the Japanese I. gracilipes, another member of the Evansia. Morphologically and ecologi­ cally, the three species, I. cristata, I. gracilipes, and I. tenuis, have much in common (see Table 1). They are all plants of moist cool shady woods where they grow in rich soil or very often in the layer of leaf mold and moss which covers the forest floor. All three possess long, slender, wiry which creep through the leaf mold and over the rocks, rooting only sparingly. One of the striking characteristics which these three species have in common is the unique manner in which they produce their flower stems away from the fan of leaves. The three species are all deciduous and in the spring three small growths will usually be found at the ends of the branches of the rhizomes. The center growth comes from the base of the past season's fan of leaves and from it emerges the flowering stem devoid of leaves except for those on the flower stem itself. The two or more side growths give rise to slender rapidly growing rhizomes which later produce fans of leaves some little distance from the flowering stem. This is very well shown in figure 4. Because the area at the base of the fan of leaves is somewhat larger in size than the diameter of the rest of the , old rhizomes often present a beaded or knotted appearance, the enlarge­ ments representing previous seasons' terminal growths. The species of the Evansia are also known as the Crested Irises because of the prominent cockscomb-like crest present on the haft of the of most of the mem­ bers of the subsection. The name Evansia was first used by Salisbury in 1812 (Trans. Hort. Soc. 1: 303) as a generic epithet for a plant earlier described by Thunberg (Trans. Linn. Soc. 2: 327, 1791) as . Feeling that his plant was suffi­ ciently distince to be given generic status, Salisbury called it Evansia chinensis. The name Evansia was first validly used as a subsection by Bentham in 1882 (Benth. and Hook. Gen. Pl. 3: 687). Included within the group at the present time are the fol­ lowing species: I. confusa Sealy, I. crista! a Soland, I. formosana Ohi, I. gracilipes Gray, I. lacustris Nutt, I. milesii M. Foster, I. pseudo-ro.rsi Chien, I. speculatrix Hance, I. tectorum Maxim, and I. wattii Baker. Whether this is phylogenetically a natural grouping or not, it is impossible to say at the present time. As Dykes ( 1913) wrote: "With the exception of the crest on the blade of the falls which is the distinguishing mark of the section there is probably no other character in which all plants here grouped together agree." Nevertheless, within the group it would appear that I. gracilipes, I. tenuis, I. cristata, and I. lacustris all possess a number of characters in common and most probably represent a natural jUNE, 1959] IRIS TENUIS 313

I !. I

f FIG. 1. Iris tenuis. a. general habit of flowering stalk; b. fan of leaves at time of flowering; c. flower from above; d. spathe valves; e. style branch; f. sepal. a, b, c, approx. X Yz ; c, d, e, f, approx. X 1. TABLE 1 I. tenuis I. gracilipes I. crista/a Leaves Ensiform, acute, pale green, scarious Ensiform, to 3 dm. long at maturity, Broadly ensiform, light green or margins to 3.5 dm. long, 1.5 cm.wide, nerves prominent yellowish-green, few sub-prominent nerves prominent nerves, to 2.5 dm.long, 2.5 em. wide Flowering Stem Slender, deeply 1-2 branched, Usually 2 branches, slender with Stem unbranched, 2-3 cauline 2-3 flowering heads, 1-2 cauline leafy at bifurcation leaves leaves, leafy bract at bifurcation Spathes Subequal, opposite, membranaceous 1 only, lanceolate, membranaceous, Broadly lanceolate, acuminate, and scarious margined 2-3 em. long, scarious opposite, equal or outer sometimes 5 mm. wide, lanceolate acuminate, shorter, 1-2-flowered 1-flowered Very slender, 0.4-1 em. long Very short 0.7-1.8 em. long, slender Perianth tube Very short, infundibuliform Short, about 1.25 cm.long, Filiform, gradually widening funnelform toward top, 4.5-6.8 cm.long Oblong spatulate, bluntly rounded Obovate, about 2.5 em. long, Obovate to spatulate, bluntly and deeply notched, white, deeply and widely emarginate, rounded at apex, to 4.5 cm ..long, sometimes bluish tinged light lavender blue 1.5 em. wide, lavender or purplish, sometimes white Crest Prominent central ridge Wavy linear crest, minute Toothed orange and white crest often yellow or yellowish fimbriate crest Style Crests 0.7 em. long, broadly obovate, erose 1 em. long, fimbriated 0.7 em. long, semi-ovate, crenate Stigma Triangular, acuminate Triangular, tongue-shaped Rounded, entire, oblong <0

_-1>-'"' z 9 N }UNE, 1959) IRIS TENUIS 315 grouping. The one character which does show considerable variation within the group is the extent of the crest which is presumably the single character that in the past has held the Evansias together. and I. lacustris both have a fimbriate crest on the haft of the sepal, I. gracilipes has a "wavy linear crest" or "prominent orange ridge," whereas I. tenuis has only a prominent ridge on the fall without any indication of its being serrate or fimbriate. That there may be some variation in the develop­ ment of the crest within a single species is indicated by the situation which exists in I. wattii. This species is known to have forms with distinctly fimbriated crests whereas other collections show only low crests which are entire. It is probable that the only form of I. watii now in cultivation is the one with the entire crests. The cytological situation within the Evansia is not sufficiently understood to permit a discussion of possible relationships within the group at the present time. Iris cristata and I.lacustris are certainly phylogenetically closely related and Dykes (1913) origi­ nally considered the latter as a variety of I. cristata but later (1924) he accorded it specific standing, a course that has been followed by later workers (Small, 1924; R. C. Foster, 193 7; Fernald, 1950), yet the chromosome numbers of the two are quite different, I. cristata having 2n=32 and I. lacustris 2n=42. The known chromo­ some numbers reported for the Evansias are listed in Table 2.

TABLE 2

Species n 2n Reported by

I. confusa (as I. wattii) 30 Simonet, 1934a Snoad, 1952 I. crista/a 12 Longley, 1928 32 Simonet, 1934a R. C. Foster, 193 7 I. formosana 28 Yasui, 1939 I. gracili pes 18 36 Kazao, 1929 Simonet, 1934 I. japonica 17, 18 Simonet, 1932 34, 54 Kazao, 1929 l. lacustris 42 Simonet, 1934 I. milesii 26 Simonet, 1932 I. speculatrix 44 Snoad, 1952 Lenz, 1959 I. tectorum 24 Sakai, 1952 28 Simonet, 1934a I. tenuis 28 Simonet, 1934a Smith and Clarkson, 1956 I. wattii 30 Lenz, 1959 The extent to which interspecific hybridization is possible with the Evansias, or the role which it may have played in the evolution of the group is unknown. Several presumed garden hybrids have been reported but only one of them has been studied cytologically, that one, Iris Paltec, which resulted from a cross between I. tectorum and a diploid tall bearded iris, is well known and is often seen in modern collections. Another hybrid of similar parentage was reported by lenz (1956) but the plants perished after blooming and before they had been studied cytologically. At the present time no hybrid is known involving Iris tenuis. Excluding I. tenuis from the discussion for the time being, the remaining members of the Evansia present an interesting example of what is perhaps the best known 316 ALISO [VoL. 4, No.2 of all discontinuous floras, that of temperate eastern Asia and temperate eastern North America. According to Fernald (1931) the identity, or close similarity, of angiospermous genera of these areas was recognized as early as 1750 by Halen, a

\ I

FIG. 2. Distribution of Iris, subsection Evansia. 1 student of linnaeus. These two widely separated geographic areas share scores of genera or subgenera in common, many of them having been pointed out by Fernald (1918, 1931). Included in the list are the tulip tree, Liriodendron tulip ifera with two living areas, one from New England to the Great lakes and south to the Gulf of Mexico, and the other (var. chinense) in China. Other examples given by Fernald include Symphlocarpus, Magnolia, Menispermum. Polyphyllum, Panax, Phryma, and Diphylleia.

'Permission to use the map shown in figures 2 and 3 was obtained from the University of Chicago Press. Goode Base Map Series. Copyright by the University of Chicago. }UNE, 1959] IRIS TENUIS 317

A less common although not unknown discontinuous distribution is that between northwestern America and southeastern United States. Here we find Pachystima with two species, one the Oregon Boxwood, P. myrsinites, distributed from Butte and Siskiyou counties in California north to British Columbia and east to the Rocky Mountains. The second species, P. canbyi, is found in the mountains of Virginia and West Vir gina and in Kentucky and .

~· \ \ "/

FrG. 3. Distribution of the genus Dicentra.

A number of genera are known with species in both temperate Asia and western North America: Included here is Boschniakia of the Orobanchaceae with B. tuberosa being found from the San Jacinto, San Bernardino and San Gabriel mountains of southern California north to British Columbia; another species, B. hookeri, is found in the coastal mountains of Washington and extends south perhaps as far as Cali- 318 ALISO [VOL. 4, No.2

fornia. In Asia B. 1'0ssica is found in Japan and B. himalaica in the Himalayan region. In the Urnbelliferae the genus Glelmia has one species, G. leiocarpa, which is found from Curry County, Oregon, north to Alaska, and another species, G. littoralis, which occurs in China, Korea, and Japan. Plant groups with disco ntinuous distributions involving the three areas of tem­ perate Asia, eastern United States and Canada, and northwestern America, although relatively rare, are known. The genus Oplopanax with three species is one example. The Dev il's Club, 0. horrida, is known from Ontario and Michigan west to Oregon and north to Alaska. Oplopanax elctltts is found in Korea and 0 . japonictts in Japan. Another genus is Dicentra with some 15 species of which D. cum/aria is found from

FIG. 4. b·is cristata. the Gaspe peninsula east to North Dakota and eastern Kansas and south to and . The variety occidentalis is found in Idaho, Oregon, and Washington. Dicentra canadensis is found from Nova Scotia to Michigan and south to North Carolina, , and Nebraska; D. chrysantha and D . f ormosana are native to western America and D. spectabilis, the well-known Bleeding Heart, is a native of Japan. Perhaps the genus most closely approximating the distribution of the Evansias (including I. temtis ) is Clintonia in the Liliaceae. In this genus C. borealis and C. ttrnbellata are found in eastern North America, C. andrewsiana and C. tmiflora in western United States, Canada, and Alaska, and C. ttdensis and C. alpina in Asia (see figure 3) . According to Axelrod (1950) the Arcto-Tertiary Flora was dominated by tem­ perate hardwood-deciduous and conifer forest species whose nearest living counter­ parts survive largely in the temperate parts of North America and Asia. In western America the closest modern equivalents are to be found in the Coast, Sierra-Cascade, and Rocky Mountain forests, whereas in eastern United States the present day plants are confined to areas of summer moisture. The Asian element comprised groups now JUNE, 1959] IRIS TENUIS 319 indigenous to eastern Asia where the climatic situation is similar to that found in eastern North America. According to Axelrod these three elements are known as having been in close association at many localities. He believes that intimate and regular association of their species suggests a rather generalized forest in the Miocene with a relatively uniform temperate climate over the lowland regions. Based on the fact that I. cristata, I. gracilipes, and I. tenttis are today plants of moist temperate forest habitats it might be postulated that the prototypes of these species were mem­ bers of this Miocene forest association which later became separated into the three distinct areas known today, each of them containing irises which have since evolved into the distinct species known today. CONCLUSIONS From the morphological, ecological, and geographical evidence presented it seems best to consider I. tenttis as a member of the subsection Evan.ria, section Apogon, of Iris, with closest affinities to I. gracilipes of Japan and I. cristata of eastern North America. It is postulated that the prototypes of these species were at one time elements found in the Arcto-Tertiary Flora of Miocene time. The nearest living counterparts of this flora are today found in temperate area forests of eastern Asia, western North America, and eastern United States, each area of which possesses one or more species of iris belonging to the Evan.ria subsection of the genus. LITERATURE CITED Axelrod, D. I. 1950. Studies in Late Tertiary Paleobotany. Carnegie Inst. Wash. Pub. No. 590. Clarkson, Q. D. 1955. Cytotaxonomic Studies in Iris, Section Apogon, Subsection Californicae Die is (Thesis). Oregon State College, Corvallis. --- 1958. Iris, section Apogon, Subsection Oregoneae, Subsect. Nov. Madrofio 14: 246-247. --- 1959. Iris, Series Oregonae, Stat. Nov. Baileya 6: 187. Dykes, W. R. 1913. The Genus Iris. Cambridge. --- 1924. A Handbook of Garden Irises. London. Fernald, M. L. 1918. The Geographic Affinities of the Vascular Floras of New England, the Maritime Provinces and Newfoundland. Am. Jour. Bot. 5: 219-236. --- 1931. Specific Segregation and Identities in Some Floras of Eastern North America and the Old World. Rhodora 33: 25-63. --- 1950. Gray"s Manual of Botany. 8th edition. New York. Foster, R. C. 1937. A Cyto-Taxonomic Survey of the North American Species of Iris. Contr. Gray Herb. No. CXIX. Lenz, L. W. 1950. Chromosome Numbers of Some Western American Plants, I. El Aliso 3: 317. --- 1956. Development of the Embryo Sac, Endosperm, and Embryo in Iris munzii and the Hybrid, I. munzii X I. sibirica "Caesar"s Brother.' El Aliso 3: 329-343. --- 1956. Two New and Unusual Iris Hybrids. El Aliso 3: 345-349. --- 1958. A Revision of the Pacific Coast Irises. Aliso 4: 1-72. Longley, A. E. 1928. Chromosome numbers in Iris species. Bull. Am. Iris Soc. No. 29. pp. 43-55. Simonet, Marc. 1932. Recherches cytologiques et genetiques chez les Iris. Bull. Bioi. de Ia France et de Ia Belgique LXXVIII, pp. 696-707. --- 1934. Nouvelles recherches cytologiques et genetiques chez les Iris. Ann. des Sci. Nat., Botanique, Ser. 10, 16: 229-383. Smith, F. H. and Q. D. Clarkson. 1956. Cytological Studies of Interspecific Hybridization in Iris, subsection Californicae. Am. Jour. Bot. 43: 582-588. Small, J. K. 1924. Manual of the Southeastern Flora. New York. Snoad, B. 1952. Chromosome counts of species and varieties of garden plants. Rep. John Innes Hort. Inst. for 1951. pp. 47-50. Starker, Carl. 1935. Iris tenuis. Am. Iris Soc. Bull. No. 55, pp. 90-91. Yasui, K. 1939. Karyological studies on Iris japonica Thunb. and its allies. Cytologia 10: 180 188.